effects of mindset on positive illusions

7/30/2019 Effects of Mindset on Positive Illusions

1/14

ATTITUDES AND SOCIAL COGNTION

Effects of Mindset on Positive IllusionsShelley E. TaylorUniversity of California, Los Angeles Peter M. GollwitzerUniversity of KonstanzS. E. Taylor and J. D . Brown's (1988) position tha t mentally healthy people exhibit positive illusionsraises a dilem ma: How do people function effectively if their perce ptions are positively biased? UsingGollwitzer's deliberative-implemental mindset distinction, we assessed whether people in a deliber-ative m indset show less evidence of positive illusions th an people in an implem ental mindset. Partic-ipants completed a mindset task and assessments of mood, self-perceptions, and perceived (invul-nerability to risk. Deliberation led to worsened mood, greater perceived risk, and poorer self-percep-tions, relative to implementation; control (no mindset) participants typically scored in between.Study 3 demonstrated that the mindset manipulation corresponds to how people actually makedecisions or implem ent th em . Results suggest that people use relatively realistic thinking when set-ting goals and m ore positive thinking when implementing them .

Taylor and Brown (1988) proposed that a mentally healthyperson is characterized not by accurate assessments ofhis or herpersonal qualities, realistic estimates of personal control, and arealistic outlook on the future but by positive illusions. Specifi-cally, they ma intained that people typically hold at least thre emildly self-aggrandizing perceptions of themselves, the world,and the future: unrealistically positive self-perceptions, an illu-sion of personal control, and unrealistic optimism about the fu-ture. T hey argued that, instead of being maladaptive, these pos-itively distorted perceptions actually foster the criteria normallyassociated with mental health: positive self-regard, the ability tocare for and about other people, the capacity for creative andproductive work, and the ability to effectively manage stress(Taylor, 1989; Taylor & Brown, 1994).Despite empirical support for the model, this portrait raisesa disturbing question: If normal people's perceptions aremarked by positive biases, how do they effectively identify andmake use of negative feedback they may encounter in the world?If people are capable of explaining away, compartm entalizing,or otherwise dismissing or minimizing negative feedback, as

Shelley E. Taylor, Depa rtmen t of Psychology, University of California,Los Angeles; Peter M. Gollwitzer, D epartm ent of Psychology, Universityof Konstanz, Konstanz, Germany.This research was supported by a grant from the National Instit ute ofMental Health (MH 42152) and funding from the Max Planck Society(Max Planck Institute for Psychological Research, Munich, Germany).We thank Lien Pham (Studies 1 and 2), Juliane Malzacher, and UteBayer (Study 3) for their assistance in collecting and analyzing the data .Brett Pelham and Gabrielle Oettingen m ade valuable comm ents on anearlier draft of the article .Correspondence concerning this article should be addressed to Shel-ley E. Taylor, Department of Psychology, 1283 Franz H all, University ofCalifornia, Los Angeles, California 90024-1563. Electronic mail maybe sent via the In ternet to [email protected] o r to [email protected].

Taylor and Brown (1988) suggested, they may not process alarge amount of useful feedback. Perhaps the self-serving illu-sions that bolster self-esteem and produce a positive mood inthe short run ultimately set people up for disappointment andfailure in the long run, precisely because the negative feedbackthey should have incorporated into their goal decisions andplanning went ignored (cf. Janoff-Bulman & Brickman, 1982;Weinstein, 1984).One potential resolution Taylor and Brown (1988) raised isthe possibility that there may be times when people are morehonest with themselves, d uring which they recognize and incor-porate negative feedback. Research on mindset by Gollwitzerand his associates (Gollwitzer, 1990, 1991; Gollwitzer, Heck-hausen, & Steller, 1990; Gollwitzer & Kinney, 1989) suggests aset of circumstances when precisely this situation may occur.Drawing on Heckhausen's (1986) theory of action phases, Goll-witzer and his colleagues argued that the predecisional phase,when potential action goals entailed by a person's desires andwishes are deliberated, is characterized by a deliberative m ind-set. Because deliberation involves the careful a ppraisal of po-tentially competing goals, the weighing of likely pros and conswith respect to each potential goal, and consideration of the fea-sibility of these goals, it is expected to foster relatively even-handed and accurate appraisal of evidence.The postdecisional phase, when people plan the implem enta-

tion of a chosen goal, is characterized by an implemental mind-set. The implemental mindset is assumed to induce participantsto muster motivation, resources, and cognitions in service ofgoal-directed actions, and thus, Gollwitzer (1990, 1991) pre-dicted, it will lead people in the direction of cognitions that fa-vor goal achievement. These cognitions may include those spe-cific to the goal-directed actions to be implem ented, as well asmore general cognitions that constitute the environment fortask implementation.To examine these arguments, Gollwitzer and Kinney (1989,Study 2) put half of their participants into a deliberative m ind-Jouma l of Personality and Social Psychology, 1995, Vol. 69, N o. 2, 213-226Copyright 1995 by the American Psychological Association, Inc. 0022-3514/95/S3.00

213


2/14

214 SHELLEY E. TAYLOR AND PETER M. GOLLWITZERset by asking them to contemp late the pro s and cons of makinga major change in their lives. The other half was placed in to animplem ental mindset. These participants were asked to plan theimplementation of a decision they ha d already made and to listwhen, where, and how they proposed to initiate goal-directedactions. Subsequent to the mindset ma nipulations, pa rticipantscompleted an ostensibly unrelated task that required them toestimate their degree of personal control. Specifically, theyworked on a contingency learning task in which they tried toturn on a light by either pressing or not pressing a button (mod-eled after Alloy & Abramson, 1979). The apparatus, in fact,produced noncontingent onset of the light. A control group per-formed the contingency task without any mindset manipula-tion. Participants in the implemental mindset condition in-ferred that they had successfully exerted personal control overthe light task when light onset was frequent, whereas partici-pants in the deliberative mindset condition did not succ umb tothis illusion-of-control effect. Moreover, Gollwitzer and Kinn eyfound that control participants (those not in either a delibera-tive or implemental mindset) showed a significantly greater illu-sion of control than deliberative p articipants b ut tended to showless of an illusion of control than im plemental participants.

These results imply tha t when people are deliberating poten-tial action goals, they are less vulnerable to the illusion of con-trol than when they plan the implementation of goal-directedactions. Do Gollwitzer and Kinney's (1989) results provide amore general context for resolving the conundrum created bypositive illusions? Tha t is, is deliberation generally a tim e whenpositive illusions are suspended, relative to implementation?There are several reasons why one cannot extrapolate directlyfrom Gollwitzer and Kinney's study to the more general phe-nomenon of positive illusions. The illusion-of-control paradigmused in Gollwitzer and Kinney's studies is highly artificial(pressing button s to turn on target lights). Because this situationis unfamiliar to participants, it may be relatively simple to affectresponses to it by means of a mindset manipulation. The otherillusions discussed by Taylor and Brown (1988), namely, self-aggrandizement and unrealistic optimism, may not be so easilyinfluenced, because people's self-perceptions and beliefs abouttheir personal futures may be m ore anchored in personal expe-rience. For example, people should be highly familiar with theirown cheerfulness, intelligence, kindness, and other personalqualities, and so such self-perceptions may not change greatlyin response to variable factors such as m indset. A second factorthat limits the generalizability of Gollwitzer and K inney's stud-ies to positive illusions more generally is the fact that Gollwitzerand Kinney addressed only uncontrollable outcomes. It is im-portant to assess whether mindset affects the perceived control-lability of outcomes that are under personal control. Third, Goll-witzer and Kinne y's investigation leaves unclear the processesby which m indset affects the illusion of control. It may be thatthe effect is a direct one, based on the cognitive contents andprocedures of a person's deliberations and thoughts related toimplem entation. Alternatively, it is possible that min dset affectsmood, which in tu rn affects positive illusions. Mood h as beenargued to be a pathway by which positive illusions affect m entalhealth (Taylor & Brown, 1988). Mood may also be the mostimm ediate consequence of circumstances th at foster positive il-lusions and, as such, may act as the day-to-day manifestation ofillusions in daily life (Taylor, Aspinwall, & Giuliano, 1993).

Thus, additional exploration of the processes by which mindsetmay affect positive illusions is warranted. Finally, Gollwitzerand K inney's studies leave unclear the question of whether peo-ple's spontaneous deliberations and implementations actuallycorrespond to the processes manipulated in the mindset manip-ulation. T hat is, do people deliberate evenhandedly, as the de-liberative mindset manipulation assumes, and, when they arefocused on implementing a task, are their cognitions specificto implementation, as Gollwitzer and Kinney's manipulationimplicitly assumes?The present set of investigations was designed to addressthese questions. On the basis of Gollwitzer and K inney's (1989)results, we argued that deliberation is a time when positive illu-sions are suspended, when people confront their talents andshortcomings and the resources and limitations of the environ-ment realistically in attem pting to decide whether a given actiongoal is fruitful and likely to be successful. As such, we predictedthat mindset w ould influence self-perceptions and the percep-tion of invulnerability to risk (unrealistic op timism) as well asthe illusion of control. By contrast, implementation may be atime when positive illusions are mustered, even exaggerated, inservice of the motivation needed to bring about goal achieve-ment. We explicitly tested these p redictions using measures ofself-perception and perceived risk that correspond more readilyto manipulations that have been used in previous studies of pos-itive illusions (Taylor, 1989; Taylor & Brown, 1988). In so doing,we sought to extend Gollwitzer and Kinney's results beyondthe laboratory paradigm in which these effects were previouslydemonstrated. In addition, we assessed the potential role ofmood as a mediator of the effects of mindset on positiveillusions.

Study 1In Study 1 we tested these predictions by manipulating mind-set and observing its effects on tasks related to positive illusions.We also assessed the impact of deliberative and implementalmindsets on mood. Pretesting of the deliberative and imple-mental mindset manipulations suggested that the deliberativemindset induced a negative mood in many participants. An as-sessment of mood enabled us to see if mood might be a media-tor whereby deliberative and implem ental mindsets affect posi-tive illusions, if such effects exist.The study also explored unrealistic optimism about risk. Atleast 120 investigations have indicated that, when asked to eval-uate their risk of succumbing to a wide variety of m isfortunesrelative to their peers, most people see themselves as less vulner-

able (Weinstein, 1993). This illusion of invulnerability is bothrobust and pervasive, encompassing a wide variety of poten-tially threatening events. We predicted that a deliberative mind-set would reduce people's tendency to see themselves as less vul-nerable than others to an array of negative events, whereas animplem ental m indset would enhance the perception that one isless vulnerable tha n others to negative events, relative to a con-trol condition.To further examine how mindset m ay affect positive illusions,we explored its impact on two dispositional measures. We in-cluded a measure of self-esteem (Rosenberg, 1965) to see if in-dividuals in a deliberative m indset experience tem porary lossesin self-esteem. Self-esteem has previously been used as a proxy


3/14

MINDSET AND ILLUSIONS 215for positive illusions in other investigations (e.g., Aspinwall &Taylor, 1992; Baumeister, Tice, & Hu tton, 1989). We also ex-amined the effect of mindset on the Life Orientation Test(LOT), a dispositional measure of optimism developed byScheier and Carver (1985), to see if mindset can induce temp o-rary shifts in expectations about the future.

MethodParticipants

Eighty-one participants (50 women and 31men) were recruited indi-vidually by student experimenters in a research methods course. Allwere college stud ents between 17 and 29 years of age, with a median ageof 21 . A classroom of 22 psychology students, 18 women and 4 men,naive as to the purpose of the experiment and of a comparable age dis-tribution as the participants in the mindset conditions, completed alldependent measures without completing the mindset tasks.1 They con-stituted a control condition for the dependent measures against whichthe deliberative and implemental mindset conditions could becompared.Procedure

Each individually run participant was approached by a student ex-perimenter, and the experiment was carried out in a quiet setting free ofdistractions. Participants were first given verbal instructions to orientthem to the experiment. They were told that the researchers were inves-tigating the process ofgoal selection and implem entation and had foundthat certain tasks help people to define and pursue their goals moreeffectively. They were told that they would be working on a predeter-mined mental exercise with respect to their personal goals that wouldenable the experim enters to evaluate these tasks further.Both the oral instructions and the written instructions followed theprocedu res of Gollwitzer and Kinney (1989); they were translated fromGerman by Gollwitzer and his associates. All translated materials andinstructions were pretested on UCLA undergraduates and revisedslightly to increase clarity and avoid confusion.2 Participants were thenrandomly assigned to either the implemental or deliberative mindsetmanipulation condition and given written instructions.Deliberative mindset condition. For the deliberative mindset condi-tion, participants were asked to name an unresolved personal problem.A brief description of these types of problems was given:

Such problems are characterized by the fact that you are not yetsure whether to take action in order to change things. You feel veryuncertain and you ask yourself whether it might not be better toleave things as they are. This means that you haven't decided totake action, bu t you have n't decided against it either.Participants were told that they should not select tasks easy to imple-ment, those for which a decision was already m ade, or those for which adecision w ould p robably never be reached. They were informed that theproblem should be complex and take the form of "Should I . . . ornot?"The analysis of the decision took three parts. Part A was concernedwith the imme diate consequences and long-term c onsequences of mak-ing a decision that involves a change. Immediate consequences werelisted first, and then long-term consequences that could ensue fromthose imm ediate consequences were listed. For each positive and nega-tive consequence, participants were asked to rate the c ertainty of occur-rence in percentage. Part B asked participants to list the expecteddifficulties th at m ight arise in trying to implem ent the decision involv-ing the change. Part C asked participants to list the immediate and long-term consequences of not making a decision, that is, of leaving things asthey are. For both positive and negative consequences, they were asked

to rate the certainty in percentage that the consequence would occur.Participants were given a sample packe t, ostensibly completed by a pre-vious participant, concerning the decision of "going on a vacation ornot." They were then instructed to work on their own unresolved per-sonal problem.Implemental mindset condition. Participants in the implementalmindset condition received a similar packet of written instructions butinstead were asked to describe an intended project. A description ofsuch a project was provided:

Intended projects are set goals which people intend to realize some-day. All intended projects have in common that people have de-cided to take action in this matter. They feel determ ined to achievethe project.Participants were told that the intended project should be complex andone that could be achieved within the next 3 months. They were askednot to invent a new project for the purpose of the exercise bu t to use onefor which they had already made a decision.After naming the project, particip ants were informed th at they wouldbe writing down the implementation steps involved in the intendedproject. They were provided with a set of questions divided into twoparts to assist them. In Part A, they were asked to list the five mostimportant steps for bringing about the project. These were describedas distinct behavioral episodes which, if successfully executed, shouldguarantee the achievement of the intended project. Part B was con-cerned with the planned execution of these steps. Participants wereasked to list when, w here, and how each step was to be performed.A sample sheet with the example of "going on vacation," ostensiblycompleted by a prior participant, was provided.Participants were askedto examine the exam ple carefully and then move on to the ma ster sheetto describe their own personal intended project.Dependent Measures

After completing the deliberative or implemental mindset task, theparticipants were given the dependent m easures. First, they completed amood scale consisting of seven adjectives tak en from the Multiple AffectAdjective Checklist (MAACL; Zuckerm an & Lubin, 1965), which hasbeen used in a previous empirical effort to assess mood (Aspinwall &Taylor, 1993). They rated themselves on each adjective using a 7-pointscale ranging from 1 (not at all) to 7 (extremely). The adjectives werehappy, downhearted, upbeat, contented, upset, sad, satisfied, lonely, anddistressed. A composite measure of mood was derived by subtractingthe summ ed self-ratings on adjectives that reflected negative mood fromthe summed self-ratings on adjectives that reflected positive mood.Next, participants completed a measure of relative perceived risk de-rived from Perloff and Fetzer (1986). Specifically, on separate items,they were asked to rate themselves and the average college student oftheir gender as to the likelihood of experiencing each offiveunpleasantfuture events. The items were: being in an automobile accident, becom-ing divorced, becoming depressed, developing a drinkin g prob lem, andbeing mugged. These ratings were completed on 7-point scales thatranged from 1 (not at all likely) to 7 (extremely likely). The m easure of

1 Because completion of the dependent m easures took approximately10 min, whereas the mindset manipulation took 30-40 min, it was de-cided, for purposes of economy, to run the control group in a groupsetting. In retrospect, this may have been a false economy, because itrendered the data collection circumstances of the control group some-what different than those in the two mindset conditions, creating ques-tions of comparability.2 For example, the phrase "change decision" translated from Germandid not make sense to participants, and so the phrase "a decision in-volving a change" was substituted.


4/14

216 SHELLEY E. TAYLOR AND PETER M. GOLLWITZERrelative risk was obtained by subtracting total self-ratings from total"average college student" ratings, thus yielding a measure of relativeinvulnerability to risk. If participants see themselves and the averagecollege student of their age and gender as equally vulnerable, then theirscores on the risk measure should both be 0; to the extent that partici-pants see themselves as more invulnerable to risk than others, the re-sulting score will be positive.

Next, participants completed the Rosenberg Self-Esteem Scale (Ro-senberg, 1965). This measure includes 10 statements reflecting self-atti-tudes (e.g., "On the whole, I am satisfied with myself"), and partici-pants indicated on a 5-point scale that ranged from 1 (not at all) to 5(very much) how much they agreed with each of the statements. The lastmeasure wa s the LOT. Participants answered eight target items and fourfiller items on 5-point scales ranging from 1 (strongly disagree) to 5(strongly agree). Sample items include the statements "If something cango wrong for me, it will" and "I'm always optimistic about my future".

After participants completed the dependent measures, the experi-menter conducted a short postexperimental interview to assess partici-pants' reactions to the experiment and to probe for suspicion. No par-ticipant expressed suspicion over the nature of the study. Participantswere then orally debriefed, given a written debriefing statement, andthanked for their participation.

ResultsManipulation Check

All deliberation and im plementation protocols completed byparticipants were inspected individually to ensure that partici-pants understood the instructions, had chosen an appropriateproject, and had completed all questionnaires. One participantin the implem entation condition had written about a project forwhich he had not yet made a decision, and consequently hisdata were not included in the analyses. The remainder of theparticipants had completed the protocols appropriately. Thosedeliberating a decision typically picked such issues as whetheror not to go to graduate school, whether or not to break off arelationship, and whether or not to move to another commu-nity. Those in the imp lementation condition typically describedthe steps involved in furthering a personal goal, such as com-pleting a degree, getting a good job, getting into graduate school,becoming involved in a relationship, or changing a personalhabit, such as losing weight or practicing a musical instrument.Preliminary Analyses

Initial inspection of the means suggested gender trends, yet itwas not possible to conduct Condition X Gender analyses be-cause of the substantial gender skew in the control condition.Therefore, we conducted a multivariate analysis of variance(MANOVA) with two between-subjects variables: mindset (de-liberative or im plemental) and gender (male or female), and fourdependent variables of mood, perceived risk, self-esteem, anddispositional optimism. We employed Wilk's criterion for thecombined ratings, and the MANOVA resulted in a significantmain effect for condition, F(4, 71)= 12.89, p < .001. However,the main effect for gender and the interaction between genderand condition were not significant (both Fs < 1). Therefore,gender was not further considered in the analyses.Dependent Measures

We performed a one-way (deliberative, implem ental, c ontrol)MANOVA on the four dep endent measures of mood , perceived

risk, self-esteem, and dispositional op timism. The use of Wilks'criterion for the combined ratings resulted in a significant maineffect, F(8, 186) = 7.42, p < .001. Mean ratings correspondingto the subsequent univariate analyses are presented in Table 1.Analysis of the m ood scale indicated that participants in thedeliberative mindset experienced significantly poorer moodthan those in the control condition and in the implementationcondition, F(2, 97) = 27.80, p < .0001. Tukey HSD post hoccomparisons revealed that the deliberative mindset mean wassignificantly different from the control and implemental mind-set means , which were not different from each other.We had predicted that people in an implemental mindset seethemselves as less susceptible to unfortunate events than thosein a deliberative mindset. This prediction was strongly sup-ported, F(2, 97) = 4.69, p < .02. Specifically, Tukey HSD posthoc comparisons revealed that implemental mindset partici-pants saw others to be at significantly greater risk for negativeevents than themselves, relative to the control group and tothose in a deliberative m indset. It should be noted that if partic-ipants showed no systematic capacity to distort their vulnera-bility to risk, one would have expected the means of all threegroups to be 0. The fact that the deliberative mindset conditionmean is 6.00 means that, even in this condition, participantsheld a moderately high degree of perceived relative invulnera-bility to risk. This suggests that, although deliberative mindsetand control participants are somewhat less unrealistically opti-mistic about their risk than implemental mindset participants,they still perceive their risk to be less than that of the averagecollege student.

We had included self-esteem in Study 1 to see if pa rticipantsin the deliberative mindset condition experienced lower self-es-teem than those in the control or implemental mindset condi-tions. This prediction also was supported, F(l, 97) = 4.72, p

effects of mindset on positive illusions

Documents